CN1836266A - Device for detecting road traveling lane - Google Patents
Device for detecting road traveling lane Download PDFInfo
- Publication number
- CN1836266A CN1836266A CNA200480023582XA CN200480023582A CN1836266A CN 1836266 A CN1836266 A CN 1836266A CN A200480023582X A CNA200480023582X A CN A200480023582XA CN 200480023582 A CN200480023582 A CN 200480023582A CN 1836266 A CN1836266 A CN 1836266A
- Authority
- CN
- China
- Prior art keywords
- mentioned
- line
- traveling lane
- curve
- lane
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/16—Anti-collision systems
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/16—Anti-collision systems
- G08G1/167—Driving aids for lane monitoring, lane changing, e.g. blind spot detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/12—Edge-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/70—Determining position or orientation of objects or cameras
- G06T7/73—Determining position or orientation of objects or cameras using feature-based methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/255—Detecting or recognising potential candidate objects based on visual cues, e.g. shapes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
- G06V10/457—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components by analysing connectivity, e.g. edge linking, connected component analysis or slices
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/56—Context or environment of the image exterior to a vehicle by using sensors mounted on the vehicle
- G06V20/588—Recognition of the road, e.g. of lane markings; Recognition of the vehicle driving pattern in relation to the road
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/04—Detecting movement of traffic to be counted or controlled using optical or ultrasonic detectors
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10016—Video; Image sequence
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30248—Vehicle exterior or interior
- G06T2207/30252—Vehicle exterior; Vicinity of vehicle
- G06T2207/30256—Lane; Road marking
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Multimedia (AREA)
- Image Analysis (AREA)
- Traffic Control Systems (AREA)
- Image Processing (AREA)
Abstract
A device for detecting a road traveling lane, capable of stably specifying a boundary line of a traveling lane. Edge point detecting means (ED) detects edge points in a contour line in an image, and segment group producing means (SD) produces line segments based on continuity and orientations of distances between edge points and produces a segment group by grouping line segments having a predetermined relationship. Further, curve detecting means (CD) detects a curve matching the segment group. Then, when segment groups forming a curve closest to the center of the traveling lane have a predetermined length and repeated cycle, lane boundary line position specifying means (LD) specifies the segment groups as the innermost marking line, and the position of the line next to the outside of the marking line is specified as the position of the boundary line of the traveling lane.
Description
Technical field
The present invention relates to the traveling lane pick-up unit on a kind of road surface, particularly relate to a kind of from continuous shooting vehicle front the road surface and detect the device for detecting road traveling lane of traveling lane the image that obtains.
Background technology
In the automatic control of automobile and driver's drive supporting etc., the traveling lane that correctly and reliably detects on the road surface from the image of taking with video camera has become very important.Usually, on the road surface with headed by the lane line on border of identification traveling lane (fare) and according to the underlined line of various purpose applications, wherein mix the different shape such just like solid line or dotted line or square shape mark line and as white or yellow the mark line of different color, further also have the situation of compound these mark lines.
For example, Fig. 3 is the example of the image DS that comprises mark line near the road surface in 2 tracks the inlet in tunnel, adopt the mark line of white or yellow solid line as the lane line LB of expression traveling lane DL left border, and with application within it the mark line of the square shape of the white of side use as the guide line LG that travels.Also have, adopt the mark line of white or yellow dotted line as the lane line RB of the right side boundary of expression traveling lane DL, and with application within it the mark line of the square shape of the white of side use as the guide line RG that travels.Usually, the width setup of these mark lines becomes 20cm, also has, and the length setting of the finish department of dashed lines labeled line becomes 8m, and the length setting of the blank parts between the finish department becomes 12m.Also have, the width setup of the mark line of square shape becomes 30cm, and the length setting of finish department becomes 2～3m, and the length setting of the blank parts between the finish department becomes 2～3m.In addition, in this application, the lane line or the guide line that travels mean the mark line of classifying from function, when the white on the face that shows the way or these yellow mark lines, are referred to as lane markings.
Device at detecting the traveling lane on the road surface of being discerned by various mark lines as described above has various motions all the time, for example disclosed content in the patent documentation 1.In this communique about vehicle road decision maker and controller of vehicle, be purpose with the stipulated standard line of correctly setting vehicle by detected many mutual adjacent mark lines, and constitute as follows.That is, detect the mark line on the road surface that is drawn in road from the photographic images that utilizes video camera to obtain, therefrom extract the mark line that become a pair of white line of dividing traveling lane.And, detect the interval of a pair of mark line that is extracted as white line.Under the situation at the interval that detects a pair of mark line that is extracted as white line, from the photographic images that utilizes video camera to obtain, detect when there are many adjacent mark lines at least one side of road, based on the interval of this constantly detected a pair of mark line as white line, will extract as white line with the most consistent a pair of mark line in this interval.
Also have, having proposed to detect lane boundary reliably in patent documentation 2 is purpose, has the lane boundary pick-up unit of following structure.That is, have: the first profile information detector, it will set higherly with respect to the sensitivity of the spatiality concentration change of view data, and extract first round profile information from view data; The second profile information pick-up unit, it will set lowlyer with respect to the sensitivity of the spatiality concentration change of view data, and extract the second outline line information from view data; The contour extraction device, it is from the first and second outline line information extraction white line group's outermost profile information, and, set the lane boundary position based on this outermost profile information.Therefore, owing to be set higherly with respect to the sensitivity of concentration change, therefore a side includes the information corresponding to the edge in gap between the white line, and the opposing party does not comprise this information, so carries out the deletion corresponding to the information at the edge in this gap easily.
And then, also proposed and above-mentioned same lane boundary pick-up unit purpose, that have following structure at patent documentation 3.That is the outermost contour extraction portion (Reference numeral 15 in the patent documentation 3.Below identical) based on the original digital image data that is stored in frame buffer part (13), with comprise outline data by the positional information at the detected edge of rim detection portion (14), extract white line group's outermost profile information.Outermost contour extraction portion (15) is based on comprising from the outline data of the positional information at the edge that original digital image data extracts, whether judge this edge corresponding to the gap that is formed between the white line that constitutes the white line group, and will from outline data, delete corresponding to the edge in this gap.
And patent documentation 4 has also proposed and above-mentioned same lane boundary pick-up unit purpose, that have following structure.That is, take the traveling lane of the moving body of the fare that comprises the regulation zone with filming apparatus, and obtain view data.View data based on this acquisition makes intensity histogram, and detects histogrammic cluster and divide into groups.And detecting in the histogram that carried out grouping becomes first middle position of histogrammic central authorities separately, and based on first middle position, detects second middle position that becomes central authorities in the histogrammic cluster that carried out grouping.And then, based on the same group histogram second middle position each other not, detect lane markings or have the lane markings group's of a plurality of lane markings central authorities, and decision lane markings boundary position, therefore, by making, can carry out the detection of reliable lane markings boundary position based on the histogrammic of view data.
On the other hand, about image processing techniques, known Hough (Hough) conversion that has as line detection method for example, just has its explanation in the non-patent literature 1 below.This Hough transformation is known as the reliable line detection method to noise, it is characterized in that, with (x, y) point transformation of coordinate system becomes (ρ, θ) in the process of the curve on the polar coordinate system, based in that (x y) is positioned at (ρ of the unique point on the same straight line on the coordinate system, θ) the curve on the coordinate system intersects on one point.And then, in recent years on computer vision, as a kind of RANSAC (Random SampleConsensus: the grab sample consistency algorithm) gazed at by the people, in the non-patent literature 2 below detailed explanation is arranged of robust method.Also have, the explanation of couple RANSAC is also arranged in the non-patent literature 3 below.
Patent documentation 1:JP spy opens the 2003-168198 communique.
Patent documentation 2:JP spy opens the 2003-187227 communique.
Patent documentation 3:JP spy opens the 2003-187252 communique.
Patent documentation 4:JP spy opens the 2003-178399 communique.
Non-patent literature 1: Tian Cunxiu is capable, and prison is repaiied " Computer Image Processing is crossed the threshold ", always grinds publication, clear and the 1st printing distribution of first published on March 10th, 60,127 pages and 128 pages.
Non-patent literature 2:Martin A.Fischero and Robert C.Bolles work " Random SampleConsensus:A Paradigm for Model Fitting with Applications to Image Analysisand Automated Cartography ", Graphics and Image Processing, vol.24 (6), the page381-395.1981 distribution.
Non-patent literature 3:Richard Hartley and Andrew Zisserman work " Multiple ViewGeometry in Computer Vision ", Cambridge University Press.2000 issues August, 101 pages to 107 pages.
Disclosure of an invention
In above-mentioned patent documentation 1, record, as the lane line of road and at least when a side detects adjacent mutually a plurality of mark line, interval based on this detected a pair of mark line of moment, to extract as white line with the most consistent a pair of mark line in this interval, the interval of the lane line of both sides will be kept necessarily as prerequisite.Also have, determining datum line from a plurality of mark lines is not easy thing, and expectation further improves.
In addition, in above-mentioned patent documentation 2, by the two kind profile testing methods different with respect to the sensitivity of spatiality concentration change, determine the outermost outline position for the gap desensitization of a plurality of mark lines, even because lighting condition etc. former thereby contrast mark line and gap is insufficient, perhaps saturated and image fault can both be determined the position of outermost profile reliably, but the extremely difficult mark line that detects the position that is positioned at original lane line.
And then, the device of record in patent documentation 3, if the interval at edge is narrow and small or the difference of the concentration of two marginal positions is less, then be identified as the gap of a plurality of mark lines, do not adopt these data and extract the outermost outline position, thereby can with the above-mentioned position of determining the outermost profile equally, reliably, but the also extremely difficult mark line that detects the position that is positioned at original lane line of this method.
And, the device of record in patent documentation 4, do that paired image carries out differential and the histogram at the edge that obtains and dividing into groups, detect the middle position of mark line separately and as the middle position of group, according to quantity of mark line etc., middle position or the most inboard position are used as datum line, still, fully satisfied the requirement of determining the position of lane markings boundary line reliably hardly.Particularly, width at the mark line of square shape shown in Figure 3 has 30cm wide, therefore if the mark line of this square shape temporarily is identified as lane line in the both sides of traveling lane, the interval that then also comprises mark line with the lane line of reality of square shape, lane width (fare width) is than the actual lane width 1m nearly that narrowed down, thereby is difficult to the control etc. of successfully travelling sometimes.Therefore, be necessary to identify reliably the mark line of square shape with respect to the boundary line of traveling lane.
Therefore, the object of the present invention is to provide a kind of from continuous shooting vehicle front the road surface and detect the device for detecting road traveling lane of traveling lane the image that obtains, can determine the position of the boundary line of traveling lane reliably.
In order to reach above-mentioned purpose, the present invention is a kind of device for detecting road traveling lane, detects traveling lane the image that obtains from being taken continuously by the filming apparatus road pavement, wherein, have: the endpoint detections device detects a plurality of marginal points in its outline line from image; The stage group implementing device, it is at by the detected a plurality of marginal points of this endpoint detections device, makes line segment based on the continuity of the range-azimuth between each marginal point, and a plurality of line segments of the relation that has regulation are carried out packetizing, thereby makes stage group; The curve detection device, it detects the curve that the stage group that makes with this stage group implementing device matches; The lane line position determining means, it will be contrasted by the stage group that is distributed near the lane boundary of the left and right sides a plurality of curves in the detected curve of this curve detection device, make with above-mentioned stage group implementing device, constituting that the stage group of the curve at the center of approaching above-mentioned traveling lane has the length of regulation and repeatedly during the cycle, be defined as the most inboard mark line, will be with respect to the center of above-mentioned traveling lane and the position of the curve adjacent with the outside of above-mentioned the most inboard mark line, be defined as the position of the boundary line of above-mentioned traveling lane.And, in above-mentioned curve, include the situation that constitutes curve in fact by a plurality of straight lines.We can say a plurality of line segments as the relation that has regulation, is the line segment at regulation, and the line segment of selecting in turn for example to be present in the scope of range-azimuth of regulation is obtained.
And in above-mentioned device for detecting road traveling lane, above-mentioned stage group implementing device in above-mentioned a plurality of line segments, will be present in other line segment in the scope of range-azimuth of regulation with respect to the line segment of regulation as a group, thereby make above-mentioned stage group.Perhaps, also can be above-mentioned stage group implementing device at based on edge point group by the detected a plurality of marginal points of above-mentioned endpoint detections device, based on the continuity of the range-azimuth between each marginal point, make above-mentioned line segment.In addition, also can be that above-mentioned stage group implementing device is in the line segment group based on above-mentioned a plurality of line segments, when having other line segments in the scope of the range-azimuth of the regulation set with respect to the line segment of regulation, judge that there is the relation of afore mentioned rules in they, they are handled as belonging to same group.
In addition, the present invention also can be a kind of device for detecting road traveling lane, detects traveling lane the image that obtains from being taken continuously by the filming apparatus road pavement, wherein, have: the endpoint detections device detects a plurality of marginal points in its outline line from image; The curve detection device, it detects the curve with the edge point group that is matched by the detected a plurality of marginal points of this endpoint detections device; The stage group implementing device, its edge point group that will be used to constitute by the detected curve of this curve detection device carries out packetizing, thereby makes stage group; The lane line position determining means, it will be contrasted by the stage group that is distributed near the lane boundary of the left and right sides a plurality of curves in the detected curve of above-mentioned curve detection device, make with above-mentioned stage group implementing device, constituting that the stage group of the curve at the center of approaching above-mentioned traveling lane has the length of regulation and repeatedly during the cycle, be defined as the most inboard mark line, will be with respect to the center of above-mentioned traveling lane and the position of the curve adjacent with the outside of above-mentioned the most inboard mark line, be defined as the position of the boundary line of above-mentioned traveling lane.
And, also can be that above-mentioned stage group implementing device is at the edge point group that is used to constitute by the detected curve of above-mentioned curve detection device, make edge histogram (edge histogram), the edge point group that will be used for histogram peak carries out packetizing, thereby makes above-mentioned stage group.And preferably above-mentioned edge histogram is made the horizontal direction edge histogram that makes in the horizontal direction with respect to the vertical composition of above-mentioned edge point group.
In addition, also can be above-mentioned endpoint detections device detect above-mentioned a plurality of marginal point on the image of being taken by above-mentioned filming apparatus after, the coordinate figure back projection of above-mentioned a plurality of marginal points to the three-dimensional pavement coordinate, is exported as above-mentioned a plurality of marginal points.
Because the present invention has said structure, therefore has following effect.Promptly, constituting that stage group near the curve at the center of traveling lane has the length of regulation and repeatedly during the cycle, be defined as the most inboard mark line, will be with respect to the center of traveling lane and the position of the curve adjacent with the outside of the most inboard mark line, be defined as the position of the boundary line of traveling lane, therefore stage group can be had the length of regulation and repeatedly the mark line of the square shape in cycle be different from the boundary line of traveling lane and foreclose reliably.Therefore, can stably determine the position of the boundary line of traveling lane.
Above-mentioned stage group implementing device passes through to constitute as mentioned above, thereby can suitably make stage group.In addition, above-mentioned endpoint detections device can correctly carry out the detection and the processing of a plurality of marginal points by constituting as mentioned above.
The simple declaration of accompanying drawing
Fig. 1 is the block diagram of primary structure of the vehicle traveling on roadway duct device of expression an embodiment of the present invention.
Fig. 2 is the block diagram of hardware configuration of the vehicle traveling on roadway duct device of expression an embodiment of the present invention.
Fig. 3 is the front view (FV) that is illustrated in an example of the image of taking in an embodiment of the present invention.
Fig. 4 is illustrated in the vertical view that is projected in a plurality of marginal points on the coordinate of road surface in an embodiment of the present invention.
Fig. 5 is illustrated in the vertical view that is projected in the line segment on the coordinate of road surface in an embodiment of the present invention.
Fig. 6 is the vertical view of an example that is illustrated in the packetizing of an embodiment of the present invention middle conductor.
Fig. 7 is illustrated in to be projected in the vertical view of the lane markings on the coordinate of road surface and the corresponding therewith histogrammic curve map of horizontal direction in other embodiments of the present invention.
The explanation of Reference numeral
The VD filming apparatus
ED endpoint detections device
SD stage group implementing device
CD curve detection device
LD lane line position determining means
The CM video camera
VB video input buffer circuit
The SY synchronizing separator circuit
The FM frame memory
The VC image processing part
VP view data control part
EP endpoint detections portion
CP curve detection portion
The SP stage group makes portion
LP lane boundary line position determination portion
The best mode that carries out an invention
Concrete embodiment at the device for detecting road traveling lane of the present invention of said structure describes with reference to the following drawings.Fig. 1 represents an embodiment of device for detecting road traveling lane, and to be taken continuously by filming apparatus VD road pavement, the mode that detects traveling lane from take resulting image constitutes.In the present embodiment, have: endpoint detections device ED detects a plurality of marginal points in its outline line from image; Stage group implementing device SD, it is at by the detected a plurality of marginal points of this endpoint detections device ED, makes line segment based on the continuity of the range-azimuth between each marginal point, and a plurality of line segments of the relation that has regulation are carried out packetizing, thereby makes stage group; Curve detection device CD, it detects the curve that the stage group that makes with this stage group implementing device SD matches.And, at lane line position determining means LD, to contrast by the stage group that is distributed near the lane boundary of the left and right sides a plurality of curves in the detected curve of curve detection device CD, make with stage group implementing device SD, constituting that stage group near the curve at traveling lane center has the length of regulation and repeatedly during the cycle, be defined as the most inboard mark line, will be with respect to this traveling lane center and with the outside position adjacent of inboard mark line, be defined as the position of the boundary line of traveling lane.
The device for detecting road traveling lane of above-mentioned Fig. 1 has hardware configuration shown in Figure 2.That is,, for example ccd video camera (the being designated hereinafter simply as video camera) CM as filming apparatus VD is installed, takes the visibility region of the vehicle front that comprises the road surface continuously in the place ahead of not shown vehicle.The signal of video signal of video camera CM carries out the A/D conversion through video input buffer circuit VB, synchronizing separator circuit SY, and is stored among the frame memory FM.By image processing part VC the view data that is stored among this frame memory FM is handled.Image processing part VC makes the SP of portion, the CP of curve detection portion by view data control part VP, the EP of endpoint detections portion, stage group and lane boundary line position determination portion LP constitutes.And the EP of endpoint detections portion, stage group make endpoint detections device ED, stage group implementing device SD, curve detection device CD and the lane line position determining means LD that the SP of portion, the CP of curve detection portion and lane boundary line position determination portion LP correspond respectively to Fig. 1.
In image processing part VC, the view data in frame memory FM, accessed the data of address designated by view data control part VP, and sent to the EP of endpoint detections portion, and detected a plurality of marginal points at this.To the marginal point data that are detected like this, in the present embodiment,, make line segment based on the continuity of the range-azimuth between each marginal point at stage group implementing device SD, a plurality of line segments of the relation that has regulation are carried out packetizing, thereby make stage group.In addition, detect the curve that matches with the stage group that makes by this stage group implementing device SD by curve detection device CD.And, at lane boundary line position determination portion LP, from as mentioned above like that by the detected curve data of the CP of curve detection portion, near the lane boundary about selection is distributed in a plurality of curves, with these a plurality of curves with make the stage group that the SP of portion makes by stage group and contrast, have the length of regulation and repeatedly during the cycle, be defined as the most inboard mark line constituting stage group near the curve at traveling lane center.And, will be with respect to this traveling lane center and the position of the curve adjacent with the outside of the most inboard mark line, be defined as the position of the boundary line of traveling lane.
The position of the boundary line of the traveling lane of Que Dinging like this, further as required, with the curvature of the width of traveling lane, road, be fed into systems control division SC (computing machine), and output to outside system equipment (not shown) via output interface circuit OU with the testing result of the position of vehicle, attitude angle etc.In addition, the CL among Fig. 2, PW, IN are respectively clock circuit, power circuit and input interface circuit.
Below, illustrate that the above-mentioned endpoint detections EP of portion, stage group make the processing of the SP of portion, the CP of curve detection portion and each one of lane boundary line position determination portion LP.At first, in the EP of endpoint detections portion, as shown in Figure 3, from the image DS that takes by video camera CM, detect a plurality of marginal points, carry out back projection to the three-dimensional pavement coordinate from the image surface (not shown) of a plurality of marginal points.That is, based on the parameter of detected a plurality of marginal points and video camera CM on image surface, as shown in Figure 4, with the coordinate figure of these a plurality of marginal points as the point group of three-dimensional pavement coordinate and carry out back projection's (line of Fig. 4 is represented the edge point group).And, because white line (LB of Fig. 3, LG, RB, the RG) color as lane markings shoals or reasons such as performance contaminated or video camera CM, cause the adjacent part of white line on the image to be connected, so shown in the top of Fig. 4, can become the edge point group different, but can not cause error by processing described later makes suitable judgement with the below part.
In the CP of curve detection portion,, for example intend drawing and contain the curve of a plurality of straight lines, thereby carry out curve fitting by above-mentioned RANSAC for by a plurality of marginal points (with Fig. 4 is representative, represents with EGP) of back projection on the road surface.Plan picture (curve fitting) about this curve can utilize above-mentioned Hough transformation, also can use for example least square method.In addition, also can carry out packetizing to edge point group EGP, carry out curve fitting based on the attribute of stipulating.
In addition, as shown in Figure 5, make the SP of portion to above-mentioned edge point group EGP, make line segment LS based on the continuity of the range-azimuth between each marginal point by stage group.Then, in the online stage group, if in the scope of the range-azimuth that certain line segment LS is set, have other line segments LS, then those line segments LS, LS are taken as and belong to same group and handle, form as shown in Figure 6 by carrying out this processing repeatedly a plurality of groups (for the center, track, with the group of inboard as SGI, with the group in the outside as SGO).And and packetized, marginal edge (on the right side of white line, representing as LS (-) among Fig. 5) is packetized too though select positive edge (in the left side of white line, representing as LS (+) among Fig. 5) in Fig. 6.
Then, do curve fitting (curve plan picture) for the line segment LS that has carried out packetizing.Also can carry out packetizing based on the attribute of regulation this moment, carries out curve fitting.Use this line segment, verification is that marginal point by which kind of attribute constitutes by the curve detection portion detected curve of CP.If for example the curve of the group SGI of Fig. 6 is made of a plurality of periodic short-terms, then this curve of decidable is that the relatively shorter mark line of intending picture square shape etc. forms.And, like this, length and cycle that line segment has regulation on vertical or horizontal, in lane boundary line position determination portion LP, when judgement is the mark line of square shape, (for example RG of Fig. 3) picks out from the lane line candidate with this curve, and judgement is the boundary line of traveling lane with respect to the curve (RB of Fig. 3) in the outside of the mark line (RG of Fig. 3) of the square shape at center, track.
Though in the above-described embodiment, at first obtain line segment LS, and its packetizing carried out curve fitting, but also can make following embodiment: shown in the dotted arrow of Fig. 1, constitute like that, detect the curve that matches with a plurality of marginal points, the edge point group that is used to constitute this curve is carried out packetizing, thereby make stage group.Promptly, in the image processing part VC of Fig. 2, detect the curve that matches with a plurality of marginal points by the CP of curve detection portion, make the vertical composition that the SP of portion is used to constitute the edge point group of this curve relatively at stage group, make the horizontal direction edge histogram, thereby will be used to constitute the edge point group packetizing of histogram peak, make stage group.
Specifically, shown in the HG of Fig. 7,, make the edge histogram of horizontal direction corresponding to by a plurality of marginal points of back projection on the road surface of three-dimensional pavement coordinate.According to this embodiment, though a plurality of peak value PK appear as shown in Figure 7, owing to contain a plurality of perpendicular line compositions at peak separately, so can constitute the edge group of each histogram peak as a group with being used to.And, length and cycle that histogram peak has regulation in vertical (or laterally), at lane boundary line position determination portion LP, when being judged as the mark line of square shape, (RG) picks out from the lane line candidate with this mark line, with respect to the center, track, with the mark line (RB) in the outside of the mark line (RG) of square shape boundary line as the track.
As mentioned above, expression is arranged in the mark line of the lane boundary on the track except simple solid line, dotted line, also existing will this simple mark line and many lines combining of the mark line of square shape, therefore in device in the past, be difficult to determine reliably and want as lane boundary and the position of detected mark line (lane line), but, in above-mentioned any one embodiment of the present invention, all can determine the position of lane line reliably.And, can realize fully satisfying the identification of the boundary line of the high reliability that warning system and control system expect.
Utilizability on the industry
Device for detecting road traveling lane of the present invention can be determined the car that travels as mentioned above like that reliably Therefore the position of the boundary line in road, can be applicable to various warning systems and control system such as vehicle etc. System.
Claims (11)
1. a device for detecting road traveling lane detects traveling lane the image that obtains from being taken continuously by the filming apparatus road pavement, it is characterized in that having:
The endpoint detections device detects a plurality of marginal points in its outline line from image;
The stage group implementing device, it is at by the detected a plurality of marginal points of this endpoint detections device, makes line segment based on the continuity of the range-azimuth between each marginal point, and a plurality of line segments of the relation that has regulation are carried out packetizing, thereby makes stage group;
The curve detection device, it detects the curve that the stage group that makes with this stage group implementing device matches;
The lane line position determining means, it will be contrasted by the stage group that is distributed near the lane boundary of the left and right sides a plurality of curves in the detected curve of this curve detection device, make with above-mentioned stage group implementing device, constituting that the stage group of the curve at the center of approaching above-mentioned traveling lane has the length of regulation and repeatedly during the cycle, be defined as the most inboard mark line, will be with respect to the center of above-mentioned traveling lane and with above-mentioned the most inboard mark line.The position of the adjacent curve in the outside is defined as the position of the boundary line of above-mentioned traveling lane.
2. device for detecting road traveling lane as claimed in claim 1, it is characterized in that above-mentioned stage group implementing device is in above-mentioned a plurality of line segments, the line segment of other that will be in the scope of range-azimuth that is present in regulation with respect to the line segment of regulation is as a group, and makes above-mentioned stage group.
3. device for detecting road traveling lane as claimed in claim 1, it is characterized in that above-mentioned stage group implementing device is at by the edge point group that constitutes with the detected a plurality of marginal points of above-mentioned endpoint detections device, based on the continuity of the range-azimuth between each marginal point, and make above-mentioned line segment.
4. device for detecting road traveling lane as claimed in claim 1, it is characterized in that, above-mentioned stage group implementing device, in line segment group based on above-mentioned a plurality of line segments, in the scope of the range-azimuth of sending out fixed at the line segment of regulation regulation during other line segments of existence, judge the relation that has afore mentioned rules, handle as belonging to same group.
5. device for detecting road traveling lane as claimed in claim 1 is characterized in that, above-mentioned curve detection device carries out curve fitting at packetized line segment, detects above-mentioned curve.
6. device for detecting road traveling lane as claimed in claim 1, it is characterized in that, above-mentioned lane line position determining means, at above-mentioned line segment in vertical or horizontal length and the cycle with regulation, and when being judged as the mark line of square shape, the mark line of this square shape is removed from the lane line candidate, with respect to the center of above-mentioned traveling lane and the curve in the outside of the mark line of above-mentioned square shape is judged to be the boundary line of above-mentioned traveling lane.
7. device for detecting road traveling lane as claimed in claim 1, it is characterized in that, above-mentioned endpoint detections device, after on the image of taking by above-mentioned filming apparatus, detecting above-mentioned a plurality of marginal point, the coordinate figure back projection of above-mentioned a plurality of marginal points to the three-dimensional pavement coordinate, is exported as above-mentioned a plurality of marginal points.
8. a device for detecting road traveling lane detects traveling lane the image that obtains from being taken continuously by the filming apparatus road pavement, it is characterized in that having:
The endpoint detections device detects a plurality of marginal points in its outline line from image;
The curve detection device, it detects the curve with the edge point group that is matched by the detected a plurality of marginal points of this endpoint detections device;
The stage group implementing device, its edge point group that will be used to constitute by the detected curve of this curve detection device carries out packetizing, thereby makes stage group;
The lane line position determining means, it will be contrasted by the stage group that is distributed near the lane boundary of the left and right sides a plurality of curves in the detected curve of above-mentioned curve detection device, make with above-mentioned stage group implementing device, constituting that the stage group of the curve at the center of approaching above-mentioned traveling lane has the length of regulation and repeatedly during the cycle, be defined as the most inboard mark line, will be with respect to the center of above-mentioned traveling lane and the position of the curve adjacent with the outside of above-mentioned the most inboard mark line, be defined as the position of the boundary line of above-mentioned traveling lane.
9. device for detecting road traveling lane as claimed in claim 8, it is characterized in that, above-mentioned stage group implementing device, at the edge point group that is used to constitute by the detected curve of above-mentioned curve detection device, make edge histogram, the edge point group that will be used for histogram peak carries out packetizing, thereby makes above-mentioned stage group.
10. device for detecting road traveling lane as claimed in claim 9, it is characterized in that, above-mentioned lane line position determining means, in above-mentioned histogram peak in vertical or horizontal length and the cycle with regulation, and when being judged as the mark line of square shape, the mark line of this square shape is removed from the lane line candidate, with respect to the center of above-mentioned traveling lane and the curve in the outside of the mark line of above-mentioned square shape is judged to be the boundary line of above-mentioned traveling lane.
11. device for detecting road traveling lane as claimed in claim 8, it is characterized in that, above-mentioned endpoint detections device, after on the image of taking by above-mentioned filming apparatus, detecting above-mentioned a plurality of marginal point, the coordinate figure back projection of above-mentioned a plurality of marginal points to the three-dimensional pavement coordinate, is exported as above-mentioned a plurality of marginal points.
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP331356/2003 | 2003-09-24 | ||
| JP2003331356A JP3956926B2 (en) | 2003-09-24 | 2003-09-24 | Road lane detection device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN1836266A true CN1836266A (en) | 2006-09-20 |
| CN100452093C CN100452093C (en) | 2009-01-14 |
Family
ID=34373039
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNB200480023582XA Expired - Fee Related CN100452093C (en) | 2003-09-24 | 2004-09-22 | Device for detecting road traveling lane |
Country Status (6)
| Country | Link |
|---|---|
| US (1) | US20070084655A1 (en) |
| EP (1) | EP1667085A4 (en) |
| JP (1) | JP3956926B2 (en) |
| KR (1) | KR100784307B1 (en) |
| CN (1) | CN100452093C (en) |
| WO (1) | WO2005029440A1 (en) |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102968770A (en) * | 2012-11-30 | 2013-03-13 | 华为技术有限公司 | Method and device for eliminating noise |
| CN101567086B (en) * | 2009-06-03 | 2014-01-08 | 北京中星微电子有限公司 | Method of lane line detection and equipment thereof |
| CN104648397A (en) * | 2013-11-19 | 2015-05-27 | 沙漠科技股份有限公司 | Lane departure warning system and method |
| CN104978562A (en) * | 2014-04-08 | 2015-10-14 | 阿尔斯通运输科技简易股份公司 | Method for detecting rails on which railroad vehicle is traveling |
| CN106494399A (en) * | 2015-08-31 | 2017-03-15 | 现代自动车株式会社 | Vehicle and the lane detection method for vehicle |
| CN107851390A (en) * | 2015-08-04 | 2018-03-27 | 日产自动车株式会社 | Step detection means and step detection method |
| CN108335404A (en) * | 2018-02-07 | 2018-07-27 | 深圳怡化电脑股份有限公司 | Edge fitting method and money-checking equipment |
| CN112384962A (en) * | 2018-07-02 | 2021-02-19 | 日产自动车株式会社 | Driving assistance method and driving assistance device |
| CN113212352A (en) * | 2020-02-06 | 2021-08-06 | 佛吉亚歌乐电子有限公司 | Image processing apparatus and image processing method |
Families Citing this family (30)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP4365352B2 (en) | 2005-07-06 | 2009-11-18 | 本田技研工業株式会社 | Vehicle and lane mark recognition device |
| JP4905648B2 (en) * | 2006-02-21 | 2012-03-28 | 学校法人東京理科大学 | Stimulus presentation device for vehicle and stimulus presentation method |
| KR101035761B1 (en) * | 2006-07-06 | 2011-05-20 | 포항공과대학교 산학협력단 | Image Processing Method and System for Lane Recognition |
| JP2009237901A (en) * | 2008-03-27 | 2009-10-15 | Zenrin Co Ltd | Method of creating road marker map |
| US8384776B2 (en) * | 2009-04-22 | 2013-02-26 | Toyota Motor Engineering And Manufacturing North America, Inc. | Detection of topological structure from sensor data with application to autonomous driving in semi-structured environments |
| JP5363921B2 (en) * | 2009-08-31 | 2013-12-11 | 富士重工業株式会社 | Vehicle white line recognition device |
| JP4968412B2 (en) * | 2010-01-29 | 2012-07-04 | トヨタ自動車株式会社 | Road information detection device and vehicle travel control device |
| JP5469509B2 (en) * | 2010-03-31 | 2014-04-16 | パナソニック株式会社 | Lane position detection device and lane position detection method |
| WO2012089261A1 (en) * | 2010-12-29 | 2012-07-05 | Tomtom Belgium Nv | Method of automatically extracting lane markings from road imagery |
| JP5452518B2 (en) * | 2011-02-09 | 2014-03-26 | 富士重工業株式会社 | Vehicle white line recognition device |
| US9098751B2 (en) * | 2011-07-27 | 2015-08-04 | Gentex Corporation | System and method for periodic lane marker identification and tracking |
| KR20130015746A (en) * | 2011-08-04 | 2013-02-14 | 엘지전자 주식회사 | Apparatus for detecting lane and method thereof |
| KR101295077B1 (en) * | 2011-12-28 | 2013-08-08 | 전자부품연구원 | Lane Departure Warning System |
| CN103150905B (en) * | 2013-02-06 | 2015-12-09 | 广州畅通智能交通科技有限公司 | Trackside installs the method for ripple detecting device detection frequently traffic flow |
| JP6169366B2 (en) * | 2013-02-08 | 2017-07-26 | 株式会社メガチップス | Object detection device, program, and integrated circuit |
| DE102014210411A1 (en) * | 2013-09-06 | 2015-03-12 | Robert Bosch Gmbh | Method and control and detection device for plausibility of a wrong-way drive of a motor vehicle |
| CN103630122B (en) * | 2013-10-15 | 2015-07-15 | 北京航天科工世纪卫星科技有限公司 | Monocular vision lane line detection method and distance measurement method thereof |
| JP6274936B2 (en) * | 2014-03-25 | 2018-02-07 | ダイハツ工業株式会社 | Driving assistance device |
| US9321461B1 (en) * | 2014-08-29 | 2016-04-26 | Google Inc. | Change detection using curve alignment |
| JP6530685B2 (en) * | 2015-09-15 | 2019-06-12 | 株式会社デンソーアイティーラボラトリ | Object detection apparatus, object detection system, object detection method and object detection program |
| RU2725561C2 (en) * | 2016-03-24 | 2020-07-02 | Ниссан Мотор Ко., Лтд. | Method and device for detection of lanes |
| US10275666B2 (en) | 2016-03-24 | 2019-04-30 | Nissan Motor Co., Ltd. | Travel lane detection method and travel lane detection device |
| JP7024176B2 (en) * | 2016-09-27 | 2022-02-24 | 日産自動車株式会社 | Track detection method and track detection device |
| KR101910256B1 (en) * | 2016-12-20 | 2018-10-22 | 전자부품연구원 | Lane Detection Method and System for Camera-based Road Curvature Estimation |
| US10860868B2 (en) * | 2018-04-18 | 2020-12-08 | Baidu Usa Llc | Lane post-processing in an autonomous driving vehicle |
| JP2020085788A (en) * | 2018-11-29 | 2020-06-04 | 太陽誘電株式会社 | Method and device for calculating iron loss |
| CN112950740B (en) * | 2019-12-10 | 2024-07-12 | 中交宇科（北京）空间信息技术有限公司 | Method, device, equipment and storage medium for generating high-precision map road center line |
| CN111611862B (en) * | 2020-04-22 | 2022-09-09 | 浙江众合科技股份有限公司 | Curve fitting-based semi-automatic labeling method for subway rail |
| KR102499334B1 (en) | 2021-06-28 | 2023-02-14 | (주)뷰런테크놀로지 | Method of detecting lane using lidar sensor and a traffic lane detection device performing method |
| KR20230071556A (en) * | 2021-11-16 | 2023-05-23 | 현대자동차주식회사 | Apparatus for controlling driving of vehicle and method thereof |
Family Cites Families (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5359666A (en) * | 1988-09-28 | 1994-10-25 | Honda Giken Kogyo Kabushiki Kaisha | Driving way judging device and method |
| JP3169483B2 (en) * | 1993-06-25 | 2001-05-28 | 富士通株式会社 | Road environment recognition device |
| JP3721594B2 (en) * | 1995-03-15 | 2005-11-30 | 日産自動車株式会社 | Road shape estimation device |
| US5904725A (en) * | 1995-04-25 | 1999-05-18 | Matsushita Electric Industrial Co., Ltd. | Local positioning apparatus |
| US5991427A (en) * | 1996-07-31 | 1999-11-23 | Aisin Seiki Kabushiki Kaisha | Method and apparatus for detecting a lane on a road |
| JP3993259B2 (en) * | 1996-07-31 | 2007-10-17 | アイシン精機株式会社 | Image processing device |
| JPH1067252A (en) * | 1996-08-29 | 1998-03-10 | Aisin Seiki Co Ltd | Vehicle running condition detector |
| JP3429167B2 (en) * | 1997-09-13 | 2003-07-22 | 本田技研工業株式会社 | White line detector for vehicles |
| JP3373773B2 (en) * | 1998-01-27 | 2003-02-04 | 株式会社デンソー | Lane mark recognition device, vehicle travel control device, and recording medium |
| JP3463858B2 (en) * | 1998-08-27 | 2003-11-05 | 矢崎総業株式会社 | Perimeter monitoring device and method |
| JP2001101415A (en) * | 1999-09-29 | 2001-04-13 | Fujitsu Ten Ltd | Image recognizing device and image processor |
| JP2001134769A (en) * | 1999-11-04 | 2001-05-18 | Honda Motor Co Ltd | Object recognizing device |
| CN1351317A (en) * | 2000-10-27 | 2002-05-29 | 新鼎系统股份有限公司 | Image detection system and method |
| JP3635244B2 (en) * | 2001-05-16 | 2005-04-06 | 富士通テン株式会社 | Curve R correction method and apparatus |
| JP3822468B2 (en) * | 2001-07-18 | 2006-09-20 | 株式会社東芝 | Image processing apparatus and method |
| JP4016735B2 (en) * | 2001-11-30 | 2007-12-05 | 株式会社日立製作所 | Lane mark recognition method |
| JP3662218B2 (en) * | 2001-12-18 | 2005-06-22 | アイシン精機株式会社 | Lane boundary detection device |
-
2003
- 2003-09-24 JP JP2003331356A patent/JP3956926B2/en not_active Expired - Fee Related
-
2004
- 2004-09-22 EP EP04787985A patent/EP1667085A4/en not_active Withdrawn
- 2004-09-22 CN CNB200480023582XA patent/CN100452093C/en not_active Expired - Fee Related
- 2004-09-22 US US10/572,956 patent/US20070084655A1/en not_active Abandoned
- 2004-09-22 KR KR1020067004529A patent/KR100784307B1/en not_active IP Right Cessation
- 2004-09-22 WO PCT/JP2004/013802 patent/WO2005029440A1/en active Application Filing
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101567086B (en) * | 2009-06-03 | 2014-01-08 | 北京中星微电子有限公司 | Method of lane line detection and equipment thereof |
| CN102968770A (en) * | 2012-11-30 | 2013-03-13 | 华为技术有限公司 | Method and device for eliminating noise |
| CN104648397A (en) * | 2013-11-19 | 2015-05-27 | 沙漠科技股份有限公司 | Lane departure warning system and method |
| CN104648397B (en) * | 2013-11-19 | 2017-05-17 | 沙漠科技股份有限公司 | Lane departure warning system and method |
| CN104978562A (en) * | 2014-04-08 | 2015-10-14 | 阿尔斯通运输科技简易股份公司 | Method for detecting rails on which railroad vehicle is traveling |
| CN107851390A (en) * | 2015-08-04 | 2018-03-27 | 日产自动车株式会社 | Step detection means and step detection method |
| CN106494399A (en) * | 2015-08-31 | 2017-03-15 | 现代自动车株式会社 | Vehicle and the lane detection method for vehicle |
| CN106494399B (en) * | 2015-08-31 | 2021-03-09 | 现代自动车株式会社 | Vehicle and lane detection method for vehicle |
| CN108335404A (en) * | 2018-02-07 | 2018-07-27 | 深圳怡化电脑股份有限公司 | Edge fitting method and money-checking equipment |
| CN112384962A (en) * | 2018-07-02 | 2021-02-19 | 日产自动车株式会社 | Driving assistance method and driving assistance device |
| CN112384962B (en) * | 2018-07-02 | 2022-06-21 | 日产自动车株式会社 | Driving assistance method and driving assistance device |
| CN113212352A (en) * | 2020-02-06 | 2021-08-06 | 佛吉亚歌乐电子有限公司 | Image processing apparatus and image processing method |
Also Published As
| Publication number | Publication date |
|---|---|
| JP2005100000A (en) | 2005-04-14 |
| EP1667085A4 (en) | 2007-02-07 |
| CN100452093C (en) | 2009-01-14 |
| EP1667085A1 (en) | 2006-06-07 |
| KR100784307B1 (en) | 2007-12-13 |
| WO2005029440A1 (en) | 2005-03-31 |
| US20070084655A1 (en) | 2007-04-19 |
| JP3956926B2 (en) | 2007-08-08 |
| KR20060057004A (en) | 2006-05-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN1836266A (en) | Device for detecting road traveling lane | |
| CN100424724C (en) | Device for detecting road traveling lane | |
| US10970566B2 (en) | Lane line detection method and apparatus | |
| US6493458B2 (en) | Local positioning apparatus, and method therefor | |
| US7046822B1 (en) | Method of detecting objects within a wide range of a road vehicle | |
| JP6519262B2 (en) | Three-dimensional object detection device, three-dimensional object detection method, three-dimensional object detection program, and mobile device control system | |
| CN102227758B (en) | Camera device | |
| CN109635737B (en) | Auxiliary vehicle navigation and positioning method based on road marking line visual recognition | |
| KR960042482A (en) | Object observation method and object observation apparatus using the method, and traffic flow measurement apparatus and parking lot observation apparatus using the apparatus | |
| CN101030256A (en) | Method and apparatus for cutting vehicle image | |
| KR102491527B1 (en) | Detection of objects in camera images | |
| JP3629935B2 (en) | Speed measurement method for moving body and speed measurement device using the method | |
| JP2000163694A (en) | Method and device for recognizing car number | |
| CN114677658A (en) | Billion-pixel dynamic large-scene image acquisition and multi-target detection method and device | |
| JP2662583B2 (en) | On-vehicle distance detection device | |
| JP2557350B2 (en) | Speed measurement method | |
| KR20240081060A (en) | Method for correcting visual sign of the object in image and apparatus thereof | |
| CN118103881A (en) | Method and device for determining the position of a vehicle | |
| JP2000182184A (en) | Method and device for detecting antenna on vehicle | |
| JP2021056841A (en) | Teacher data forming device and image classification device | |
| JP2003317105A (en) | Travel path recognition device | |
| JPH11167624A (en) | Method for recognizing white line on road | |
| JPH0331999A (en) | System for processing traffic information |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| C17 | Cessation of patent right | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20090114 Termination date: 20091022 |